'Internet plus ' approach to mapping exposure and seismic vulnerability of buildings in a context of rapid socioeconomic growth: a case study in Tangshan, China

摘要

This paper presents the development of an "Internet+" approach to mapping exposure and seismic vulnerability of buildings in a context of rapid socioeconomic growth. This approach is a combination of the following interdependent components: (1) extraction of footprint areas of a large number of buildings from high-resolution Google Earth images; (2) estimation of floor numbers of these buildings with an integrated use of high-resolution Google Earth images, Tencent/Baidu Street Views, crowdsourcing data, and associated building-relevant local knowledge; and (3) identification of structural types of these buildings by a combined use of crowdsourcing data and associated building-relevant local knowledge. The efficacy of this "Internet+" approach was demonstrated through an application in Tangshan, China. Field-based verification indicated that the overall mean absolute percentage error of the proposed "Internet+" approach in assessing the total floor area of the addressed buildings was 4.64 . The verification also showed that the overall consistency between the estimated structural types using the proposed approach and the actual structural types of the buildings with structural type uncertainties could reach 97.54 , with a kappa coefficient of 0.94. Because of its good accuracy, noteworthy speed, substantial labor savings, negligible cost and distinctive capability in covering large areas in near real time, this "Internet+" approach might have promising prospects in actual seismic loss risk reduction challenges.